Based on the analysis of extinction, absorption, scattering cross sections, asymmetry factor parameters and particle size distribution function of several typical kinds of haze particles, the transmission and scattering characteristics of 0.55 μm laser in haze media with different components and different mixing modes are studied by Monte Carlo method, and the effects of several typical kinds of pollutant particles on light transmittance are also compared. The results show that at the same concentration, the transmittance in ammonium sulfate medium is the highest and that in carbon aerosol medium is the lowest, and the transmittance in fog is less than that in haze. The comparative analysis of the transmittance in the medium composed of cluster particles in different mixed modes shows that the attenuation of light by typical pollutant particles in external mixing mode or in internal mixing mode is almost the same, while the distribution of water components in the medium has an obvious effect on the transmittance of laser beam.

Lidar can quickly realize non-invasive measurement of atmospheric wind field and obtain accurate wind vectors. In order to verify the feasibility of turbulence observation with lidar and obtain the turbulent characteristics, an observation experiment was carried out in Shenzhen, China, by using coherent Doppler lidar. According to the principle of Reynolds decomposition, wavelet decomposition is used to obtain the turbulent pulsating and analyze the turbulent motion characteristics. The results show that daily average turbulence intensity in Shenzhen presents a “single peak and single valley” structure, which is correlated with the temperature change during the observation. The vertical transportation caused by turbulent kinetic energy is mainly concentrated after 12:00 in the daytime and the correlation coefficient between turbulent kinetic energy and turbulent dissipation rate is 0.77. The turbulent power spectral density is basically consistent with Kolmogorov “-5/3” law in the inertial sub-range. The research results verify that the wind lidar can estimate the turbulence parameters accurately.

Based on multiple data such as synoptic data, HYSPLIT, lidar and wind data, a severe sandstorm process in Gansu Province, China, from March 26 to 28, 2019 is comprehensively analyzed, and its occurrence, development, temporal and spacial variation, vertical structure features as well as the source and transport path of the sandstorm. The results show that, (1) the horizontal trough at upper level is the key factor causing the sandstorm. The sandstorm originated from the Taklimakan Desert and affected Gansu Province from west to east, resulting in serious air pollution in several downstream cities. (2) The dust transport path reached the north of Gansu Province, and there was obvious sedimentation during the eastward movement of the dust, resulting in the continuous reduction of dust concentration in the process of transport. It is shown that the analysis of dust transport time can provide reference for the air quality prediction in downstream cities. (3) For the semi-arid desert areas, strong dust may also occur even when the wind speed is small. The results of this study indicate that the ground-based lidar products can not only be used to analyze the vertical development of the sandstorm, but also act as a good indicator for the transport path and source of the sandstorm.

Based on the data of temperature, humidity, cloud height, visibility and all-sky view imaging from the automatic meteorological system of Hefei Xinqiao Airport, China, the infrared characteristics of all kinds of clouds as well as the evolution of clouds and fog over the airport from July 2019 to April 2020, are analyzed. The results show that: (1) the higher the cloud bottom height is, the lower the temperature is, which means that the higher clouds have the weaker thermal radiation resulting in the darker infrared cloud image, and vice versa. And different types of clouds have their own unique infrared characteristics. (2) As for the average shortest diameter of the largest cloud block of wave cloudsat various heights, that of stratocumulus is 2-4 times that of altocumulus, and that of altocumulus is 2-4 times that of cirrocumulus. Generally, fractostratus shows fluffy-cotton alike structures, while fractonimbus are usually in the form of smoke band or scattered sheet. (3) There is a clear transformation process between cloud and fog. Radiation fog generally occurs during the period when clouds just disperse and at least half of the sky is clear, and the fog gradually rises to clouds and continuesto dissipate. When a breach is formed in the sky, it indicates that the fog begins to dissipate, which can be used as an important indicator for the rapid recovery of visibility. The evolution of the non-convective precipitation cloud layer shows that the cloud base height changes from high to low, and the cloud layer changes from dispersed texture to uniform layer at the same time. The uniform cloud layer always appears in the early stage of the precipitation, and the relative humidity of ground presents a sharp increase.

Based on the comprehensive analysis of the correlation between visibility and various meteorological elements, the short-term prediction model for highway visibility was studied by using the visibility, temperature, humidity and wind strength data from Anhui Provincial Highway Visibility Observation Station Network. Experiments were conducted with humidity, temperature, average wind speed, instantaneous wind speed, and maximum wind speed as the input layer of the BP neural network, and the visibility as the output layer. The results show that the overall deviation of the experimental data is within the acceptable range. For sequential test samples, the tests whose relative error within 20% account for 68.6% of the total number of tests. In each random sample test, the BP network simulation output shows a good correlation with the test sample, with a correlation coefficient between 0.6 and 0.8. The low-visibility random sample test results show thatthe root mean square error between the model output value and the sample value is in the range of 700-850 m, with little amplitude of variation, indicating that the neural network algorithm has relatively high stablility.

Sulfur dioxide (SO2) and sulfur trioxide (SO3) are important substances in waste gas emission, and do great harm to both the environment and human's health. However, in-situ and real-time measurement of SO2 and SO3 in the process of emissions has always been a challenge. Herein, tunable diode laser absorption spectroscopy (TDLAS) technology is presented to determine the mass of SO2 and SO3 simultaneously using a quantum cascade laser emitting at 7.16 μm, and the sensitivity and robustness of the measurement system is improved by employing wavelength modulation spectroscopy. In this work, a simple single-pass gas cell with low-volume is used at high temperature and low pressure to ensure the spectra of SO2 and SO3 can be separated completely when measured simultaneously. Therefore, the concentration measurement of both SO2 and SO3 are reliable, since their spectral lines are not affected byspectral overlapping due to broadening. In addition, the influence of temperature on SO2 concentration measurement is corrected, and it is proposed to calibrate the concentration of SO3 with the standard concentration SO2 gas. According tothe Allan deviation analysis, the minimum detection limit for SO2 can reach 1.98×10-6 cm3·cm-3 with the integration time of 34 s, and that of SO3 is estimated as 1.575×10-6 cm3·cm-3. The rise time and the fall time of the system are rapid as 16 s and 18 s, respectively.

Aiming at the problem that the ultraviolet-visible (UV-Vis) spectroscopy water quality detection system is susceptible to the noise interference from the instrument itself and the external environment,and there are a large number of system and stray light noise in the measured spectrum data, based on the analysis of the noise source of the water quality detection system by UV-Vis spectroscopy, a denoising method using genetic algorithm to optimize wavelet threshold is proposed and compared with the wavelet soft threshold, SG smoothing and median filtering methods. In order to evaluate the denoising effect, denoising experiments are carried out on the UV-Vis spectrum data of potassium hydrogen phthalate standard solution with the same concentration. The genetic algorithm is used to select the wavelet optimal threshold for denoising, and at the same time, the traditional wavelet soft threshold denoising, SG smoothing denoising and median filter denoising are also used for comparison. In order to verify the practical feasibility of the algorithm, the four methods are further used to denoise the spectra of actual water sample from a domestic sewage and outlets of a sewage treatment plant. The experimentalresults show that the wavelet threshold denoising effect based on genetic algorithm is obvious. Compared with the traditional wavelet soft threshold denoising, SG smoothing and median filtering methods, the signal-to-noise ratio of the new denoising method has been improved by 2.2994, 5.7066 and 2.6155 dB respectively, the root mean square error has been increased by 0.0028, 0.0087 and 0.0033, and the peak signal-to-noise ratio has increased by 2.0837, 5.2569 and 2.7375 dB. It is shown that the wavelet threshold denoising based on genetic algorithm not only suppresses the noise in spectral data, but also improves the system accuracy, which provides a new solution for water quality spectral denoising by ultraviolet-visible spectroscopy.

Based on MODIS long term observations of aerosol daily product MOD04-L2, the temporal and spatial variations of aerosol optical properties and aerosol types in China from 2004 to 2018 are obtained using linear tendency estimation and AOD-AE aerosol classification methods. It is shown that during 2004-2018 : (1) high aerosol optical depth (AOD), values (550nm) are distributed in large urban agglomerations with low altitude, dense population and developed industries, while low AOD values are distributed in sparsely populated areas with high vegetation coverage mountains and grasslands. Areas such as Guizhou and the edge of Sichuan Basin are highlighted with high ngstrm exponents (AE) and northwestern desert area with low values. (2) In 73 percent regions of China, AOD shows a decreasing trend during the observation period, and AE in the Eastern part of the “Hu Huanyong Line” also shows a decreasing trend. Especially, it is found that both AOD and AE decreased significantly during 2014-2018. (3) In terms of seasonal and monthly change trends, AE and clean continental aerosol are opposite to AOD, while urban industrial aerosol is the same as AOD. (4) The proportion of clean continental aerosols has increased year by year since 2014, indicating that air quality in China has gradually improved.